Well, like the title initially asks, can you really “bulletproof” an Atari AR-II board? From what I have found on the Internet, and older material like Star Tech Journals, the answer really is no. Pretty much all sources that mention “bulletproof” and “AR-II” on the same page are only taking about performing the Sense mods, and nothing else.
A small amount of pages mention replacing parts (caps, regulators, etc.) due to their age, but that is not bulletproofing either – that is just a good idea. But I think I can actually bulletproof it, and maybe other power supplies as well.
So I was helping a newly minted arcade game owner when his first purchase, which is his favorite arcade game: Tempest. Shortly after getting it the X-Size pot fell apart and the game ended up only showing a Multicolored spot (larger than a just a “dot”) which changed depending on what the game was doing. We soldered in a larger replacement (dangling from some wires) but while this did change the display a little when the pot was articulated, it did not fix the problem.
He had joined a FB arcade repair group and asked for help there and ended up with ideas ranging from his brightness being too high (it was not – I think the person that responded thought the game was drawing normally but with retrace lines and a white dot in the middle, kind of like how it looks when you’re drawing on an oscilloscope with no Z input), voltages too low (they were not), and one even going down the path of impending AR-II failure(!) and going on about replacing leaky sense resistors and the bottlecap transistor. Oy vey!
Got to work on a busted Centipede that booted to a white background screen that briefly showed the start of the attract mode (things drawn in green or purple) before crashing, watchdogging and doing it all over again. Test mode would do the same thing as well.
Disabling the watchdog (by grounding the WDDIS test point) did not help as the game would just essentially do the same things but just not reset after it crashed.
Going through my boards, I came across a second Dig Dug board. This one boots to static garbage and no sound. Watchdog is barking.
Removing one of the CPUs gets different garbage. Tried replacing with a known good one, no change.
Checking the pins shows me that a couple of the data lines, from D3 to D6 (I cannot remember which ones right now) are completely floating! Rarely see that. On a hunch, I replace the CPU’s socket, but this changes nothing.
Board traces go two a few chips (a 7474 and some other multiplexers or demultiplexers), and the floating lines seem to follow them. But following the traces is tricky, so I gotta dig up some schematics to see what the Hell is going on.
My previous post, about Cocktail #1, clearly documents the “easy one.” This second one is gonna be another story. It has a very dead monitor, and a non-functional board. I have not yet benched the monitor, but it has no neck glow, which is never a good sign.
The board has problems, but at least I know it is not the CPU as I already swapped it, and I know that it is not a power problem on the second Cocktail, because the board has the same behavior in Cocktail #1.
Swapped all socketed chips to no effect — they all work on the other board. Power points seemed right on the board, but I put it into my working cabinet to verify… and while doing so I managed to cross GND with a +5v trace and smoked R30 quite nicely on my AR-II board.
I have replaced it, but am still getting voltages that are too high. Gonna have to hunt other likley suspects…
As far as the board goes, I am gonna have to break out the Fluke 9010 to work on this one… Right after building an adapter for it so I cam power it up on the bench.
The monitor I should know more about once I have it on the bench and take a good look at it while I am capping it (and doing the sync upgrade).
Turns out that this one had a cracked flyback. I ordered a replacement kit (flyback and caps) from The Real Bob Roberts and replaced the flyback (my first flyback replacement!). Have not yet put it back into the game and fired it up, though. Maybe after winter…
Well, I ended up taking two Centipede cocktails off of another local collector as a package deal. Still want that Gorf, though… Will have to come back to that one soon!
But I digress… So I have two Centipede cocktails here. One of them has a dead monitor (and a non-working board, I later discovered), and the other will not sync:
(Note how dirty the control panel is, as well as the color of the button and trackball to compare with another picture later in this post.) Checked the wiring, the connectors, adjustments/controls and nothing would get it right, so I went to capping it. Here is a picture of it half way done on my bench:
After capping it (and also doing the sync improvement upgrade while I was there), all it took was a few adjustments and the screen came up sharp and clear. This image also shows one of the rebuilt trackballs, which is now running as smooth as when it came out of the crate:
The text looks a bit blurry, but only in the picture. The screen is really sharp! You can also see a new shiny trackball at the bottom too. Next to it is a really dirty button. Here is one of the panels (from the other cocktail, because its panels are in better condition) with a replacement button and a (unmounted) rebuilt trackball after being cleaned up a bit:
All in all, not bad for about 3 or so hours of work, if I do say so myself. (3 hours for everything, not just cleaning this panel! 🙂
Now, that other cocktail… that one is gonna be a bit harder…
Problem: Stiff spinning steering modules and hard shifting between gears
(Quick little story – while bringing this beast into its location, I fell and got pinned under it for a few minutes. It is amazing how much having your chest compressed effects your ability to call out for help! 🙂
OK – onto the next issue… Its steering wheels were pretty stiff, and if you tried to give them a good hard spin they would come to a halt within a couple of turns. Removing and disassembling them was easy. Turns out the problem was that the old grease in them had coagulated/thickened so that it was about as viscous and sticky as cold honey!
Getting the grease out of the barrel and off of the shaft and sleeves literally took ~10 minutes (for each module) with some rubbing alcohol, elbow grease and rags. I had a friend helping me (Sean) and we each tackled one sterring wheel each.
After getting that old gunk out, a quick application of some light lithium grease on everything solves that problem – the wheels will now spin for at least 10 turns easily.
For the shifter modules, I just applied some powered graphite to the shifter “ball” at the opening of the shifter and after a few shifts to get it all around, the shifts are much easier now.
Solution: Cleaned and lubricated the steering modules, lubricated the shifter modules.
Note: One of the steering modules has a broken shaft/cone, which caused the wheel to be off-center, and it was being held in place strictly by the force of the retaining bolt that goes through the entire assembly. I managed to get it a bit straighter than it was before, but it still is broken internally. Not sure if I want to go through sanding down the two halves to try to get them melded or epoxied together – I am afraid of shortening the cone too much and causing problems. The wheel works and the game is playable, so I might just leave it as it is.
Bringing home a non-working Atari/Kee Games Sprint 2 and getting it running
Problem: Static garbage on screen
Purchased a Sprint 2 in non-working condition. Saw a picture of it, monitor worked – it showing what appeared to be static garbage on it. Got it home and confirmed that yes, it is showing static garbage (first image below). Turning it off for a couple of seconds and then back on again keeps pretty much the same garbage display, but sometimes gets some sounds out of it (engine and/or screeching sounds). Moving the self-test switch does nothing. First suspect that CPU is not running, because the screen is static (i.e. not changing while the game is on).
After checking the usual suspects (voltages, loose wires, harness/wire burns because this is an older Atari game…), I yank the CPU to see if I get the same results (to narrow it down to the CPU). This causes a different effect, a screen filled with a single character and I get sound, so I presume for now that the CPU is working and the problem lies elsewhere.
Atari Kangaroo – After watching the attract mode for a little bit (~30 seconds), the game started to show some visual artifacts on the RHS of the screen. They were horizontal blurbs that looked like multicolor static. However, each of the blurbs, while in a random vertical location, were identical.
I bought two or three sets of unknown/untested/bad boards sometime last year. I have recently started going through them to determine the severity of their problem(s). (This is taking a surprisingly long amount of time due to the time it takes to create a JAMMA adapters for each unique interface! :/ )
Anyhoo, I came across a complete, intact Kangaroo board last weekend! After creating a minimal adapter (Power and Video), the board came right up! I tried its self-test and it passed. After watching the attract mode for a little bit (~30 seconds), the game started to show some visual artifacts on the RHS of the screen. They were horizontal blurbs that looked like multicolor static. However, each of the blurbs, while in a random vertical location, were identical.
Purchased described as “Has a ROM error“. Started by creating a JAMMA adapter, and then read the EPROM at 2E, it was not found under Romident, burned and installed a replacement. No change. Removed and read all ROMs in row/column 2, and all but one of them failed to be identified by Romident, burned and installed 5 more replacements, working! Still have to wire up an amplifier to correctly test the sound, but looks good so far.
[07/30/03] I finally found the brain cell that got me to realize that I could use my inductive listener (a telecom tool, often used with a tone tracer) to test the sound. It works!
Solution: Replaced bad ROMs
(Recently, the board is starting to develop an intermittent problem with a bad socket. I am going to replace it as the first use of my soon-to-be-here combination soldering/desoldering station.)